Wednesday, December 22, 2010

The amazing truth about PISA scores: USA beats Western Europe, ties with Asia.


What I have learned recently and want to share with you is that once we correct (even crudely) for demography in the 2009 PISA scores, American students outperform Western Europe by significant margins and tie with Asian students.
Jump to the graphs if you don't want to read my boring set-up and methodology.

The main theme in my blog is that we shouldn’t confuse policy with culture, and with demographic factors.

For instance, education scholars have known for decades that the home environment of the kids and the education levels of the parents are very important for student outcomes. We also know that immigrant kids have a more difficult time at school, in part because they don’t know the language.

Take me as an example. The school me and my brother attended was in a basement in Tehran, had no modern resources, and largely focused on religious indoctrination. But we had a good home background. Our father attended a college in the west a few years (our mother didn’t, despite stratospheric scores test scores, because at the time you didn’t send a good Kurdish girl to another city to study). So we did well in school. Conversely, the first few years in Sweden I had bad grades, in part because I didn’t master the language.

The point I am trying to make is that the school in Sweden was objectively superior to the school in Iran. But I scored lower in Sweden, because of factors outside the control of the education system. If you want to compare the effect of the school, you have to isolate those external factors and make an apples-to-apples comparison.

However, this is not at all how the media is presenting the recent PISA scores. For example there is a lot of attention of the score of the kids in Shanghai, the according to the NYT is supposed to “stun” us or something.

It's dumb to compare one of the most elite cities in a country with entire nations, and to draw policy-inference from such a comparison. Shanghai has 3 times the average income of China! It is also naive to trust the Chinese government when they tell us the data is representative of the entire nation. Either you compare Shanghai to New York City, or you compare the entire country of China, including the rural part, with other large nations. Most of the news and policy conclusions we read about PISA-scores in the New York Times is thus pure nonsense.

1. Correcting for the demography:


In almost all European countries, immigrants from third world countries score lower than native born kids.

Why? No one know exactly why. Language, culture, home environment, income of parents, the education level of the parents and social problems in the neighborhood and peer groups norms are among likely explanations. But it is generally not true that the schools themselves are worse for immigrants than natives. In welfare states, immigrants often (thought not always) go to the same or similar schools and have as much or likely more resources per student.

So the fact that immigrant students in mixed schools do worse than Swedish kids used to a few decades ago in homogeneous schools does not it out of itself prove that Swedish public schools have become worse.

Of course, the biggest myth that the media reporting of PISA scores propagates is that the American public school system is horrible.

The liberal left in U.S and in Europe loves this myth, because they get to demand more government spending, and at the same time get to gloat about how much smarter Europeans are than Americans. The right also kind of likes the myth, because they get to blame social problems on the government, and scare the public about Chinese competitiveness.

We all know that Asian students beat Americans students, which "proves" that they must have a better education system. This inference is considered common sense among public intellectuals. Well, expect for the fact that Asian kids in the American school system actually score slightly better than Asian kids in North-East-Asia!

So maybe it’s not that there is something magical about Asian schools, and has more to do with the extraordinary focus on education in Asian culture, with their self-discipline and with their favorable home environment.

There are 3 parts to the PISA test, Reading, Math, and Science. I will just make it simple and use the average score of the 3 tests. This is not strictly correct, but in practice it doesn’t influence the results, while making it much easier for the reader. (the reason it doesn't influence the results is that countries that are good at one part tend to be good at other parts of the test.)

The simplest thing to do in order to get an apples-to-apples comparison is to at least correct for demography and cultural background. For instance, Finland scores the best of any European country. However first and second generation immigrant students in Finland do not outperform native Swedish, and score 50 points below native Finns (more on this later).

On PISA, 50 points is a lot. To give you a comparison, 50 points is larger than the difference between Sweden and Turkey. A crude rule of thumb here is that 50 points is 0.5 standard deviations.

The problem is that different countries have different share of immigrants. Sweden in 2009 PISA data had 17%, and Finland 4%. It’s just not fair to the Swedish public school system to demand that they must produce the same outcome, when Sweden has many more disadvantaged students. Similarly schools with African-American students who are plagued by racism, discrimination, crime, broken homes, poverty and other social problems are not necessarily worse just because their students don’t achieve the same results as affluent suburbs of Chicago. In fact, the most reliable data I have seen suggests that American minority schools on average have slightly more money than white schools. It’s just that the social problems they face are too much to overcome for the schools. It is illogical to blame the public school system for things out of its hands.

So let’s start by removing those with foreign background immigrants from the sample when comparing European countries with each other. I define immigrants here as those with a parent born outside the country, so it includes second generation immigrants. This is fairly easy for Europe.

In the case of America, 99% of the population originates from other countries, be they England, Italy, Sweden, India, Africa, Hong-Kong or Mexico. If we want to isolate the effect of the United States public school system, we should compare the immigrant groups with their home country. For those majority of Americans whose ancestors originate from Europe, we obviously want to compare them with Europe. For some groups, such as Indians, this is inappropriate. The reason is that mainly the most gifted Indians get to migrate to America to work or study.

However, as I have argued previously, there is strong reason to believe that this problem of so called biased selection does not apply to historic European migration to the United States at the aggregate level. The people who left Europe were not better educated than those who stayed. Immigrants were perhaps more motivated, but often poorer than average.

So similar to my comparison of GDP levels, let us compare Americans with European ancestry (about 65% of the U.S population, and not some sort of elite) with Europeans in Europe. We remove Asians, Mexicans, African-Americans and other countries that are best compared to their home nations. In Europe, we remove immigrants.

The results are astonishing at least to me. Rather than being at the bottom of the class, United States students are 7th best out of 28, and far better than the average of Western European nations where they largely originate from.




The mean score of Americans with European ancestry is 524, compared to 506 in Europe, when first and second generation immigrants are excluded. So much for the bigoted notions that Americans are dumb and Europeans are smart. This is also opposed to everything I have been taught about the American public school system.

For Asian-American students (remember this includes Vietnam, Thailand and other less developed countries outside Northeast Asia), the mean PISA score is 534, same as 533 for the average of Japan, South Korea, Singapore and Hong Kong. Here we have two biases going in opposite directions: Asians in the U.S are selected. On the other hand we are comparing the richest and best scoring Asian countries with all Americans with origin in South and East Asia.


2. Policy-Implications


Libertarians in the United States have often claimed that the public school system (which has more than 90% of the students) is a disaster. They blame this on government control and on teachers unions. However, it is completely unfair to demand that a public school in southern California where most of the students are recent immigrants from Mexico whose parents have no experience in higher education (only 4% of all Mexican immigrates have a college degree, compared to over 50% of Indian immigrants) should perform as well as a private school in Silicon Valley.

The libertarians have no answer why European and Asian countries that also have public school systems score higher than the United States (unadjusted for demography). Top scoring Finland has strong teacher unions, just as California.

Similarly, the left claims that the American education system is horrible, because Americans don’t invest enough in education. The left has no answer when you point out that the United States spends insanely more than Europe and East Asia on education. According to the OECD, the United States spends about 50% more per pupil than the average for Western Europe, and 40% more than Japan.



Another policy implication is that Europe can learn from American public schools, which appear to be better than most European countries. I can only compare Sweden with the U.S, but I can tell you that from my experience, the American system is superior. I always thought this was just anecdotal evidence, but I am beginning to realize that American schools are indeed better.

For example, we don’t have any real equivalent to Advanced Placements classes. We have cheaper and worse textbooks. The teachers on average have far less education. I could go on.

Nor is it any longer a mystery to me why Americans spend so much more on education and (falsely appear) to get out less in output.

But of course the biggest implication is that most Europeans and all American liberals have lost the bragging right about their side being smarter than Americans.

3. Immigrant PISA scores compared to natives


This is again the mean difference of the 3 parts of PISA.


Australia is the only country with a negative gap, which means Australian immigrants actually score better than natives. Canada is similar. The Australian-Canadian skill based migration system is at work here, generating less inequality (even short term).

The other pattern appears to be that the gap is almost constant in the remaining Western European countries. This may be important to keep in mind, whenever people claim that uniquely Swedish policies are causing poor immigrant educational outcomes.

127 comments:

  1. Thanks.

    The lesson I take from the last graph showing smaller native-immigrant score gaps in the Anglosphere is that English-speaking countries can recruit ambitious immigrants precisely because they are English-speaking countries. Say you are an ambitious parent in Seoul or Shanghai or Bangalore who would like your child to have the possibility to immigrate to a more prosperous country: which second language are you going to have your kids study? Probably not Danish or Italian, right?

    ReplyDelete
  2. I have being saying this for years. It’s quite obvious when you compare Iranian immigrants who are middle aged or above. In Sweden, many or perhaps most never master Swedish.

    In the U.S, they can at least work with the high-school/movie English that most of them know.

    ReplyDelete
  3. Really nice work.

    Was it straightforward to get the disaggregated information (native ethnics versus 1st/2nd generation immigrants) for the countries whose scores you graphed?

    ReplyDelete
  4. This was a nice post. I don't see anyone else doing these country of origin comparisons and they are an important contribution.

    However, as a product of the public school system and with former public school teachers among my family and friends, I can tell you there are serious organizational problems with the urban public school systems of the United States.

    Without denying that that parental culture and educational attainment have some effect on educational outcomes, we can't let badly run urban schools off the hook that easily.

    ReplyDelete
  5. AMAc:

    Not really, although in 2009 no longer have data on the public database for first generation immigrants alone (they do for 2006).

    There should be a link to the database.

    I prefer first and second, so it was not a major problem.

    There is also micro-data you can download, if someone wants to sink their teeth’s into this.

    One:

    I am only saying that American Schools on average appears no worse than European schools, not that any of the two public schools systems is optimal.

    ReplyDelete
  6. Excellent work. We in Oz try fairly hard to cherry-pick our migrants: it seems to be working.

    One can also see this in the "bangs for buck": we spend significantly less per student than the US but get similar results.

    We also have one of the largest private school sectors in the OECD (about a third of all students), and it is growing. So we have considerable internal competition between schools. The US has competition between states and cities, with a relatively mobile population -- that may make a difference.

    ReplyDelete
  7. In this post, you stress environmental factors for differences in achievement. Earlier, you have written that IQ is highly hereditary. What's your opinion of the importance of IQ differences between ethnic groups?

    ReplyDelete
  8. In my case I came to the US as a 6 year old. I consistently scored at the 75th percentile in grammar school. My children scored at the 90th to 99th percentile. You do not need to be native born to learn adequate English as this can be accomplished in a 2/3 years. My wife's parents were both born in Europe and came to the US when they were in their 30's. There is no need to exclude immigrants from your table as the language can be learned rather quickly and easily.

    ReplyDelete
  9. Bert,

    There is sufficient evidence (such as twin studies) and a scientific consensus that the variability in IQ within a country is in part genetic.

    There is however no evidence about race difference in IQ and a consensus among scientists that no race differences in IQ have been proven.

    Thus I always operate under the assumption that there are NO RACE DIFFERENCES IN IQ, and that differences in school outcome between countries and groups are entirely due to culture, investment in education, social problems etc.

    I hope you will notice that my attempts to use a “country fixed effect” methodology has the advantage that it completely sidesteps this debate (as well as many other problems, especially culture), since you just compare one group (say, Swedes in Sweden ) under system X with the *same* group (Swedes in America) under system Y.

    europeasant

    “There is no need to exclude immigrants from your table as the language can be learned rather quickly and easily.”

    I would disagree. Sure, certain people, such as you, can learn another language perfectly in a few years. Many others cannot, unless they move to the new country when very young.

    My Swedish isn’t perfect, and will never be. My English is even worse. In the Swedish equivalent of the SAT, I had a perfect score on all parts except Swedish vocabulary, where I made several mistakes.

    And having grown up among immigrants, I have seen plenty of people who did well in their own country struggle with mastering or even learning the language.

    Even if only half the group does worse because of language, it depresses the mean result.

    ReplyDelete
  10. ***a consensus among scientists***

    When privately surveyed
    I think the results are a bit different. Although it's probably sensible to avoid that debate for present purposes.

    ReplyDelete
  11. "Rather than being at the bottom of the class, United States students are 7th best out of 28"

    Not exactly. Accordingly to your data US and Belgium are in 6th place.

    Much more importantly, only one large country, Germany, outperform US and only by 1 pnt.

    Canada and Netherlands are middle-sized counties outperforming US by 6 and 3 pnts respectively.

    The rest of the top are small countries and not directly comparable.

    All in all, an excellent results for the team USA and another giant ruling class lie is exposed.

    ReplyDelete
  12. Excellent point.

    If you divide a large unit such as Europe or the U.S into smaller parts, like Europe, the variance in outcome goes up.

    Small states such as Massachusetts, New Hampshire, Connecticut and Iowa score close to half a standard deviation above the national U.S average in education. They would surely beat most of the European countries (except perhaps Finland, which really does amazingly well).


    For example, in per capita income, Norway beats the U.S. However, D.C, Delaware, Connecticut, NY, Massachusetts, Alaska and New Jersey beat Norway. And the U.S as a whole beats Western Europe as a whole.

    http://super-economy.blogspot.com/2010/01/dynamic-america-poor-europe.html

    I am not aware of PISA-scores by state (but I could probably do TIMSS by state if people like).

    Preferably either you compare small European outliers with U.S states, or you compare Western Europe as a whole with the U.S as a whole.

    The left loves the cherry picking though. They simply assume that if American adopts European policy, America will achieve social utopian outcomes as tiny outliers such as Scandinavia/ Holland/Switzerland.

    But what about the poor performance of Italy and Spain, that alone have 5 times more people that all of Scandinavia?

    What about the mediocre performance of England, Germany and France? Don’t forget that Germany includes East Germany.

    Or for that matter, what about the tiny outliers at the bottom such as Scotland, Greece, Wales and Portugal?

    It is more likely that moving the U.S towards Europe will move American outcomes (generally downward) towards the mean Western European outcome, rather than magically to the small, atypical utopias of the north.

    ReplyDelete
  13. How did you correct for demography for the USA scores? Can you please explain your method? I am genuinely curious as to how you got those numbers corrected.

    As for Asian-Americans scoring higher than Asians in NE Asia, I think this can be easily explained, at least partially, by the fact that Asian immigrants are largely selected for income and educational qualifications. I'm a Singaporean of Chinese descent and most Singaporean immigrants in the US come from a much higher socio-economic stratum than the average Singaporeans. The same is probably true of Chinese (Taiwan, HK and PRC) immigrants to the US. They are in no way representative of the home populations in NE Asia.

    ReplyDelete
  14. The American numbers is simply excluding the 35% of the population that doesn’t originate from Europe, when making a comparison to Europe.

    As I wrote, many Asians are children whose parents were selected based on ability. On the other hand, and as I again wrote; only half of Asians-American are from North-East Asia (Japan-Korea and various Chinese). The rest are from poorer countries such as the Philippines, Vietnam, Laos, Malaysia, Thailand, Indonesia.

    Those guys are not elite, many are even refugees. And while not as visible, they are as nomerous as the rich Asians.

    If we were comparing Japan-Korea-Hong-Kong-Singapore with *only* Asian-Americans from those countries and China, the American advantage would be even greater.

    Or, if we compare Asian-Americans with all of Asia, not only the 4 rich countries above as I did, the American advantage would be larger still. Thailand as an example got 422 on the PISA. I didn't do this, because I think it makes no sense to compare rich countries with poorer countries, and thus unfair to Asia.

    But this doesn't mean we should ignore this second bias, which actually appears to be a little bigger than the first one.

    My comparison is therefore if anything a little too generous to Japan-Korea-Hong-Kong-Singapore.

    ReplyDelete
  15. Very interesting post. I'm an American conservative with an adoptive black-American son, so education policy is very important to me. I used to support American public schooling until law school. When I started to examine the system, it began appearing less and less democratic and more an instrument of social control (which is exactly what Prussians designed it to be before American intellectuals brought the system back to America). But that is neither here nor there in this discussion of PISA data.

    One difficulty I have with this analysis is why certain groups (specifically black-Americans) are excluded from the European-origin group. The answer may appear obvious (they're of African descent), but that shouldn't necessarily exclude them from the European group. The majority of modern Black-Americans came here centuries ago as slaves. Since that time, they have become less and less African and more an more European. Much of predominant black-American culture still has trappings of Africanism, but they are much more like white Americans than black Africans. Hence, excluding them from the European-origin group and placing them in the African-origin group seems incorrect as a matter of methodology. I may be completely off base here. Please tell me if I am.

    (Aside: data in America demonstrate black-African immigrants score better on educational tests and have better educational and vocational outcomes than black-Americans. This might be a result of selection bias, but it is what it is.)

    ReplyDelete
  16. Great blog! I'm an American who has lived about 13 years in several East Asian countries, and taught for about 6 years in Japan. I also spend a goodly amount of time in the UK, and have substitute taught in a dozen or so public schools in the US.

    I tend to agree: American schools are not that bad. This goes against the grain of my conservative biases, but hey, daffy as I may see their politics, I think most the teachers I interact with do a pretty good job.

    And that doesn't even consider the quality of American universities. I remember talking with one of my Japanese bosses . . . We agreed that the best would be a combination of Japanese system for the bad students, and the American system for the best students. But the real problem with bad American students seems to be a rotting culture -- not the education they are provided. It's almost like a caste system, or like the tribes in H. G. Wells' Time Machine, in one high school.

    In mainland China, I concur. There's no way rural schools are turning out students of the quality of schools in Shanghai. Even college kids in Nanjing want to run off to Shanghai after they graduate.

    But anyway, thanks for setting the record straight.

    ReplyDelete
  17. Great Article. It actually comes together and makes sense.

    I just want to point out a small discredit in your write up. I see your English isn't perfect, and nor is your understanding of American political parties.

    Libertarians are equal to Far Right Conservatives.
    Liberals are Left side Democrats.

    I think you mixed up the two words, Liberals and Libertarians. Libertarians actually cry out for less government, less spending, people don't need to be babysat by the government and can think for themselves. Liberals on the other hand want more government spending, more money thrown into public schools and unconsciously think the government is there to babysit the people.

    The main reason I "think" liberals want to poor money into the public school system, is because most teachers/educators are liberals. If there was a study of politcal parties in the American Education system, it would vastly weigh towards most educators are Liberals/Democrats.

    ReplyDelete
  18. Tino:
    Excellent post.

    This makes me feel a bit better about the American education system.

    However, I am still a bit bothered by the fact that Asian Americans only tied Asians. I understand that some of the Asian Americans are from (or their parents are from) poor countries. But Asian Americans are vastly richer than Koreans and we spend way more on education, yet we only tie them. Yes, a comparison of Korean-Americans with Koreans would be a better comparison. Similarly, it seems that the Germans get roughly the same results (assuming we think these tests actually mean a lot), but they spend less and are poorer than we are.

    At first glance, we seem to be wasting lots of money. Perhaps this money is going into things that are not tested on these exams. Another idea is that we spend more, because we have higher wages than other countries do. E.g. We might have equally good haircuts, but maybe we spend way more on haircuts. What do you think?

    ReplyDelete
  19. Spoiled,

    Thanks, and you are right about my English. If you put grammatical and spelling errors in the comments, I will fix them immediately.

    However, as a classical liberal with libertarian tendencies, I know perfectly well what the difference between an American liberal and a libertarian is.

    I criticize both groups in regard to education policy. Libertarians are too negative about the American public school system, and about American public teacher unions.

    A common libertarian claim is that school choice would fix the problem of schools in areas with lots of social problem. Careful evidence suggests that while school choice helps a little (mostly with stuff outside the school, such as getting arrested), the effect on test scores is small.

    http://ideas.repec.org/a/ecm/emetrp/v74y2006i5p1191-1230.html

    Libertarians don’t have a theory of social poverty, so they sort of just ignore it. If the only thing you have is a hammer (free markets), everything (pervasive social problems) looks like a nail.

    I agree with libertarians on many issues, such as fiscal policy and social policy. However I consider libertarianism too ideologically dogmatic. For me an ideology is just a model of the world to help us sort information, not laws handed down from god (which I and most libertarians and don't belive in).

    ReplyDelete
  20. redon,

    You make 2 excellent points.

    1. The United States has to spend a lot to beat Europe and tie with Asia.

    Overall, and like health, the relationship between education spending and test score outcomes is weak and concave.

    Sure, going from $1.000 to $4.000 per pupil makes a lot of difference. But going from $7.000 to $10.000 doesn’t do all that much.

    Even if the effect is small, you may want to pay it. Also, the main impact may not be math test scores, which depends a lot on latent ability and is hard for the school to change.

    However the richer school can focus more on “softer” things, like literature, arts, non-cognetive skills or norm formation.

    2. Most of the cost of the school system is wages to teachers and staff. The cost of high skill teachers and staff is higher in the United States than the average of Europe and much higher than Korea, because average wages as well as the skill premium is higher in the United States.

    Educated teachers are non-tradable, which means America can’t just hire Korean or Italian teachers to replace American ones.

    Thus the United States has some of the highest teacher salaries in the world. That’s not to say teacher salaries are high compare to other high skill professions in the U.S, just that they are higher than what teachers make in Europe.

    Also, let me point out that un-appreciated American teachers work more hours than European teachers, according to the OECD.

    ReplyDelete
  21. Very nice post! I have discussions with US conservatives and liberals often who want to excoriate the public school system and punish the 'bad' teachers through various means, as if the system is just awash with them. When I ask them what problem they are trying to solve with these teacher-punishment measures they don't seem to have an answer. They just feel that the US educational system is not what it should be, and it's the fault of all the 'bad' teachers.

    It's great to have this insight into our educational system. I agree that in general US teachers work their asses off doing the best they can for their students (I know a few teachers, and they all put in crazy hours to help their students), but so much of their work is to overcome problems that are primarily out of their hands (social issues), but the teachers are still blamed for it.

    In light of that, thank you for a write-up that seems very fair and shows the positive outcomes that our system and the teacher's hard work produce.

    ReplyDelete
  22. Thanks for this very interesting work. You raise a good point, and one that clearly should have been mentioned by journalists when reporting the results.

    I think you're overstating your position, however. Raw levels of educational achievement still matter. Industry doesn't care about whether the U.S. produces an amazing number of skilled scientists, relative to American demographics. Industry cares about how many skilled scientists America produces. Full stop.

    Policymakers should care, also. If lots more people are poor in America, compared to Nation X, then achieving good educational outcomes among the poor should be a higher priority in the U.S. than in X.

    I also think you rely on anecdotal evidence far too much. Your own personal experience does make a point about adjusting the numbers for population characteristics, yes. But have you ever sat in on classes in Harlem or Compton? Why should anyone place any more weight, when generalizing, on your seeming epiphany?

    ReplyDelete
  23. Hi Tino,

    Thanks for your reply.

    I will like to point out that Asian Americans of all ethnic groups (Chinese, Thai, Korean, Filipino, Japanese etc) have heavily urbanized populations. Hence, it is not fair to compare their educational attainment to those of Asian Asians. Even immigrants from the poorer Asian countries such as Thailand, the Philippines, Vietnam, etc are from the upper socio-economic strata of their home societies.

    "For Asian-American students (remember this includes Vietnam, Thailand and other less developed countries outside Northeast Asia), the mean PISA score is 534, same as 533 for the average of Japan, South Korea, Singapore and Hong Kong. Here we have two biases going in opposite directions: Asians in the U.S are selected. On the other hand we are comparing the richest and best scoring Asian countries with all Americans with origin in South and East Asia."

    In descending order of population size, according to Wikipedia, the major groups Chinese-Americans, Filipino-Americans, Indian-Americans, Vietnamese-Americans, Korean-Americans and Japanese-Americans. All groups have higher than average college graduation rates. The second bias that you mentioned is not a big effect because Thai/Malaysian/Laotian/etc Americans form a relatively small percentage of the Asian Americans population.

    ReplyDelete
  24. I think you did not correctly interpret your data. When you refer to "getting rid of demography", you are in fact getting rid of ethnic groups : the sentence "For those majority of Americans whose ancestors originate from Europe, we obviously want to compare them with Europe." is misleading.

    If you retain people originating from Europe, you're basically saying that you can study them in isolation, and this is not the case : you are comparing people in Europe who are (integrated since a long time in their country) with people in US who are (from a given ethnic group). You should compare apples to apples and compare with US students who are (integrated since a long time in their country).

    So, that could lead to a false interpretation : you could easily believe that it means being from a non-european ethnic group is a problem in the US, and try to judge the country in this view.

    In fact, I think the intepretation of your results is much more simple : if I'm not mistaken, the 35% US citizens you excluded are not the wealthiest. You are probably getting rid of the poorest part (let's say 20%, but you can apply the same thinking with any figure) of the population.
    In Europe, you keep more of the population, so you keep more poors, basically.

    Which means : you are comparing (relatively) wealthy US people with (a little) less wealthy people in Europe.

    There may be a lot of other additional bias due to the difference among the EU and US samples, of course, but I believe that one is easy to understand as an example.

    The comparison is very interesting by the way, led me to rethink all this and provided insights, thanks for the work you shared.

    ReplyDelete
  25. Quang:

    1. Even if you only care about outcome, you have to understand the causes of the outcome.

    People simply assume that the cause of the disappointing Americans mean is the school system.

    This is flawed logic, when attempting to identify what social scientists refer to as the treatment effect (the effect of the school system itself). There is a lot more that goes into the human capital production function that just the school system.

    By attempting to at least correct for one of those factors, demography, we find suggestive evidence that the American school system doesn’t seem to be the problem.

    2. As for me “rely on anecdotal evidence far too much”, not a single claim I make is based on anecdotal evidence.

    It is 100% based on very simple statistics from PISA. The anecdotes are just there to make the statistical analysis more readable, and to add context.

    So just ignore all the anecdotes, and criticize the statistical analysis. American with European origin in the American school system beat the average of Europeans with European origins in the European school system.

    That’s it. No more, no less.

    ReplyDelete
  26. Fox,

    So Singapore and Hong-Kong (or Shanghai) are
    not examples of “urbanized populations”?!?

    Even Japan is one of the most urban countries on the planet (80% urban), and has been since the 16th century.

    “he second bias that you mentioned is not a big effect because Thai/Malaysian/Laotian/etc Americans form a relatively small percentage of the Asian Americans population.”

    You claim is inaccurate. According to U.S Census selected population profiles, in 2006-2008, Americans with race group (note: not country of birth) as “Asian” were 13 million.

    Of these, All self-identified Chinese-Americans (including Taiwan, Hong-Kong and Chinese from Indonesia and other non-majorety Chinese nations) plus all Japan and all Korea constitute 39% of Asian-Americans.

    As I said, they may be more visible because of their success, but they not the majority.

    http://factfinder.census.gov/servlet/DatasetMainPageServlet?_program=ACS

    However, your concern about selection is valid. Next week I will try to break out Americans by country of origin to better test the theory.

    ReplyDelete
  27. Jo:

    Americans with European ancestry are not mainly recent immigrants. There is 200 million of them, and only a couple of million were born in Western Europe.

    I could try to remove the 20% poorest Europeans from each sample if you like, but it's not easy.

    ReplyDelete
  28. 1. Your point is well-taken. I agree that demographic factors play a role that we must analyze, in addition to the schools themselves.

    But the correction you are doing is very messy. It would be one thing if the students you exclude were evenly distributed across schools. But they're not -- they're highly concentrated in America's very worst schools, which are subsequently not captured in your results.


    2. I beg to differ that what you wrote was "100% statistics." Try this:

    "I always thought this was just anecdotal evidence, but I am beginning to realize that American schools are indeed better.

    For example, we don’t have any real equivalent to Advanced Placements classes. We have cheaper and worse textbooks. The teachers on average have far less education. I could go on."

    It is far from clear, for instance, that U.S. schools really have better textbooks on the whole (and if they do, you didn't show it). I have visited many urban schools here that barely have books at all.

    ReplyDelete
  29. 1. "they're highly concentrated in America's very worst schools"

    What do you base this claim on? The minority schools themselves have as much or slightly more money per pupil, and follow the same curriculum.

    Perhaps you mean the neighborhood the school is located in is worse, or that the children bring more problem from home to the school?

    If that's what you mean, than that is something quite different from the school itself and its teachers being worse, right?

    Especially if you are comparing the school to Scandinavian schools where the students have far less home/neighborhood/peer problems.

    2. You are completely misunderstand the text if that's your example of me using anecdotal evidence.

    What I wrote was the previously I used to think my observations about American schools was JUST anecdotal evidence, so I didn't make much of it (I didn't write it up in a blog, for instance).

    Now that I have real evidence, I am re-interpreting my observations.

    Anecdotal evidence can never substitute for systematic evidence. It can however be a pretty good complement, if you also have statistics.

    3.

    "I have visited many urban schools here that barely have books at all."

    Please tell us the names of those schools, and which school district they are in. I will look up the numbers for you.

    Going back as early as 1990, even before No Child Left Behind, the Department of Education could detect no difference in average expenditure between majority and minority schools.

    http://nces.ed.gov/pubs/97917.pdf

    Some people mentioned Hispanic Compton, one of the poorest neighborhoods in the United States. Average per pupil expenditure in Compton is $10,013, compared to the national average of $10,469.

    http://www.neighborhoodscout.com/ca/compton/schools/

    Compton spends far more per pupil than Sweden.

    Schools in parts of Sweden with lots of social problems and many immigrants spend more than the parts with all middle-class Swedes.

    Just because output differs, you can't simply ASSUME without looking up the data that the cause is money.

    Money is probably less important than work ethic, norms in support of education, having discipline in the classroom and the values of your peers.

    ReplyDelete
  30. Tino,

    can you back up your claims about substantially higher teacher salaries in the US vs Europe?

    Here's a comparison of salaries in higher education from the European University Institute:
    http://www.eui.eu/ProgrammesAndFellowships/AcademicCareersObservatory/CareerComparisons/SalaryComparisons.aspx

    Here we see that the US ranks above the median, but hardly in first place. Italy spends very little, commensurably with the test scores above. The figures for Russia and the Ukraine are nearly an order of magnitude lower! Of course, these are figures for secondary education, but some cursory googling seems to suggest that teachers in Scandinavian countries in particular are paid at least as well as in the US.

    There are several other factors accounting for the higher costs per pupil in the US that come to mind. Costs of textbooks, software, and so on come to mind -- nowhere in the world are educational materials priced as astronomically as they are here, though in truth it's difficult to say without a thorough breakdown of the budget, from federal down to the local levels.

    My issue with the main analysis is that multiple factors are being manipulated, and a substantial number of schools are effectively excluded from evaluation. Tighter controls for income, language and age of immigration would make for some interesting conclusions, I'm sure.

    Also, I'm curious why the Asian countries are missing from the adjusted chart, am I missing something? It seems that we'd be well below #7 if they were charted.

    I certainly do agree that cultural concerns (both in the families and within the school system) generally outweigh questions of straightforward funding for supplies, salaries, and so on, once a certain level of basic sufficiency is reached. Affecting the necessary cultural change to address this is incredibly complex, and it's silly to think that "the ruling class" are somehow manipulating you into giving up your tax dollars to be channeled directly into fat-cat elementary school teachers' pockets. Throwing money at the problem (which clearly exists, within some sector) is the most obvious solution, after all. Analysis of this type (but much greater depth) at the DOE level might be a step in the right direction.

    ReplyDelete
  31. This comment has been removed by the author.

    ReplyDelete
  32. 1. That is not teacher salaries, that's for university professors.

    2. According to the OECD Education At a Glance 2010 the United States would be 8th highest in terms of teacher salaries:

    PPP-adjusted, dollars per year, 2008, teacher with 15-year experience in Lower secondary.

    http://dx.doi.org/10.1787/888932310510.

    1 Luxembourg (98849)
    2 Switzerland (64580)
    3 Germany (59156)
    4 Ireland (54100)
    5 Netherlands (50227)
    6 Spain (46794)
    7 England (44630)

    8 United States (44000)

    9 Denmark (42308)
    10 Belgium (Fl.) (41093)
    11 Austria (40993)
    12 Finland (40953)
    13 Norway (37023)
    14 Portugal (35486)
    15 Italy (34331)
    16 France (34316)
    17 Sweden (33885)
    18 Greece (31946)
    19 Iceland (27226)
    20 Czech Republic (22084)
    21 Poland (16137)
    22 Hungary (15049)

    Although many European countries (such as Scandinavia) have higher teacher salaries in nominal dollars, because the dollar is so weak compared to the Europe.

    Perhaps that's what you are thinking of.

    3.
    "Also, I'm curious why the Asian countries are missing from the adjusted chart, am I missing something? It seems that we'd be well below #7 if they were charted."

    Because non-Hispanic white Americans are not from Asia, they are from Europe.

    ReplyDelete
  33. By the way, before someone complains:

    I didn't pick the reference group (15 year experience lower secondary). That is the reference group the OECD highlights, I suppose because they think that it's representative.

    "because the dollar is so weak compared to the Europe" should of course be Euro.

    ReplyDelete
  34. Yes, I was aware that that table specified university professor salaries (I'd meant to say "post-secondary" instead of "secondary" up there), but I couldn't find a similarly straightforward comparison for k-12 educators on short notice. I also was comparing dollars vs converted dollars, which is probably not quite correct.

    I think my point still stands, though; being in 8th place (and we could argue about conversion here, so maybe even slightly lower) for teacher salaries vs 2nd place for overall expenditures implies that there are overruns in other parts of the budget (there are fairly large countries both above and below us in the ranks, which implies that it's probably not a scale effect).

    Re: Asian countries, my fault, I scrolled back up and looked at the chart after writing a part of my comment and lost the correct context for it. The chart does appear to show that the US is in 7th place overall if you don't consider the context, maybe it should be titled more descriptively (the text on the chart also makes it seem like it's treating *all* the countries).

    ReplyDelete
  35. "implies that there are overruns in other parts of the budget"

    Or the number of teachers, or non-teacher staff, or supplies and structures, or payroll taxes, or a million others things.

    Spending more will just get you more of what America already has.

    ReplyDelete
  36. But you are still not comparing apples to apples. A third generation immigrant from Carribean is excluded from the US score, but included in the UK score.

    Also, EU-15 and European Americans have very different nationality distributions. For example, there are relatively few Spaniards in the European Americans group

    ReplyDelete
  37. Also, I would expect late European arrivals to be among top performers in the USA (due to cherry picking). Similar groups in Europe are excluded.

    ReplyDelete
  38. This is some awful statistics work. You say demography, but you're adjusting for race. You're not adjusting for socioeconomic status or for culture, but for... race. You're just presenting means, too -- no SD? How can we tell what the spread is within each group?

    Take your politics out of your analysis, and hit the statistics books. This is poor work used to further an agenda rather than arrive at a better representation of the truth.

    ReplyDelete
  39. So, are African-Americans considered "immigrants" in your analysis?

    ReplyDelete
  40. You have, in the comments, done extensive extra analysis when presented with pertinent questions. Why have you refused to do such analysis, or even respond to, the questions regarding your findings on African-Americans?

    ReplyDelete
  41. "A third generation immigrant from Carribean is excluded from the US score, but included in the UK score."

    "late European arrivals"

    1. This is both true, and would be a problem, except that in both cases it's too quantitatively small to matter.

    Remember, we are comparing two continents, one with 200 million and one with over 300 million people.

    There has been too little immigration from Western Europe to America recent years to matter for a sample of 15-year olds. Less than 1% of American whites are European immigrants, and those are mostly older people.

    As for third generation, think about it.

    You are testing a 15-year old. For that person to be a third generation immigrant, with each generation on average being 25 years, their ancestors have to have been in the country in 1945. Say 1950, if they had kids early.

    In 1950, there were very few non-European permanent residents in Western Europe.

    According to a source on Wikipedia, in 1951 there were 15000 black Caribbean in the U.K. That's 0.03% of the population, in one country. I doubt 0.03% will significantly affect the results.

    It's not enough to nit-pick and raise theoretical concerns, the issues you point to must be quantitatively important to potentially alter the results.

    2. "Also, EU-15 and European Americans have very different nationality distributions. For example, there are relatively few Spaniards in the European Americans group"

    Good point. I have done this once before (for GDP), so I considered doing it. Since you ask, let's do it now:

    Instead of weighing each European group based on their share of the European population, I will weight them based on their contribution to Western European-American ancestry.

    So Ireland for instance is only a little more than 1% of Europe, but 18% of Western European-American ancestry.

    Here is the data http://1.bp.blogspot.com/_-EMpadQx4hM/S00PAivXUEI/AAAAAAAAAAk/bangxl04qZk/s1600-h/3.png (the source is U.S Census).

    Anyway it turns out that if Americans performed exactly as native Europeans their country of origin, so if those with Swedish-American ancestry would score 505 and those with Irish-Americans 501, and so on, the score of this virtual-Europe would be 508.5.

    That's slightly higher than 506 (the actual European score), but still significantly below the actual score of kids with European origin under the American System (524).

    ReplyDelete
  42. As for the person calling this "awful statistics work",

    since this is an serious accusation, let's go step by step.

    1. "You say demography, but you're adjusting for race."

    First, race is a sub-category of demography, genius.

    Second, even if you object to doing analysis using data on race for some ideological reason, that is hardly a criticism of my " statistics work", is it?

    More about race later.

    2. " You're not adjusting for socioeconomic status or for culture,"

    Yes, that's exactly what I am doing.

    Ethnic origins and culture align almost perfectly. It's more likely than someone whose ethnicity is Mexican-American has Latin culture, than an fourth generation German-Irish American living in Kansas.

    If I want to separate all non-European cultures, the only way to do this is to remove non-European ethnic origins.

    3. " You're just presenting means, too -- no SD? How can we tell what the spread is within each group?"

    First, I wrote approximately what the S.D was. It's there in the text. Don't waste my time if you are too lazy to read the text.

    Second, if you really care about the S.D of the individual countries, why don't you just click on the link I provided and go find out? (the S.Ds are fairly similar for most countries, nothing interesting there).

    I didn't waste readers time on the S.D, because I am comparing two continents, so the sample size becomes so large (and the standard error so small), that the large differences in means we observe are apparently statistically and economically significant.

    The difference between U.S and Europe is 18. That's about 5 times larger than the standard error for non-Hispanic whites in the United States. That's all the information you need.

    (for western Europe, the sample sizes are tens of thousands, so the bounds are even tighter).

    4. "hit the statistics books."

    I certainly need to learn a lot more statistics. But between the two of us, I am not the one having difficulty following a simple exercise at an under-gradate level.

    ReplyDelete
  43. I think you are sorely mistaken about race and culture being equivalent. We're part of a multi-university research initiative on culture -- if it were just race, our life would be a lot easier!

    Regarding statistics, of course you're going to get "significant" differences at very high sample sizes. And presenting the distribution of these scores (beyond the standard error that you reference) is rather meaningful. Since we're working with adjustments, you definitely want to know what the shape of the distribution is! Especially within the groups themselves. Knowing the standard error for all the groups together doesn't tell us much.

    When you start with a flawed definition for the construct you're adjusting for, your results are not exactly meaningful.

    As a scientist, you should care more about uncovering the truth of the matter than bending the discussion to focus on political agenda. There's substantial questions about the method you follow starting with step 1.

    ReplyDelete
  44. Libertarians don't ignore social poverty, they have been clearly demonstrating that people solve these problems themselves without policy saviors when given the freedom to do so. Elinor Ostrum's Nobel Prize winning work is just one example.

    Libertarian critiques of public schools are not that vouchers yield better outcomes this year, but that any step towards decentralizing control over schools will lead to the DISCOVERY of better solutions at the micro level. A few big thinkers pulling the levers on an enormous national school system guarantee that truly new and innovative solutions will never be found.

    ReplyDelete
  45. Actually, 1950 is date of birth cut off and not arrival cut off. Arrival cutoff is more like 1970-1975 which significantly increases the size of the group(at least tenfold according to wikipedia). And what other groups are we missing?
    A huge number(over 1 million according to wikipedia) of arabs/berbers immigrated to France in 60s/early 70s. Their grandkids could very well be around 15 now.
    Sweden had a lot of immigration in 50s and 60s from poorer countries. Their grandkids could very well be around 15 as well.
    I would expect similar issues to affect many other countries.
    I don't believe adjusting for those problems would eliminate the difference completely, but it would lower it by a couple of points.

    Also, even if the numbers are correct, they don't necessarily mean that the schools are better. An alternative theory would be good influence from German(and other "smart") immigrants on Italian(and other "stupid") immigrants leading to higher scores for the "stupid" groups. Do you have american data by background to confirm why american numbers are higher?

    ReplyDelete
  46. "Sweden had a lot of immigration in 50s and 60s from poorer countries."

    Nope. This is a myth. Sweden had no quantitatively relevant immigration from outside of Europe until the 1980s. They had a few thousand Turks. The "poor" nations you are referring to are Finland, the Baltic's and Italy.

    I repeat to you: This is a quantitative issue. Third generation non-European immigrants are very few, far less than 1% of the population.
    How do I know this? Because second generation non-European immigrant, something we have plenty of data on, is still surprisingly small.

    Third generation is a fraction of second generation, which is already very small.
    I don't need data on all American groups, if I have data on all European countries of origin and weight accordingly. That takes care of your "smart German" problem.

    Rodney:

    Some libertarians, such as Oster and Hayek, care about and understand these issues. Many other, such as followers of Ayn Rand, as well as anarchy-capitalists, do not.

    Also, Oster doesn't guarantee that social problems are solved without a state, she just shows that sometimes it happens. What about all the times it doesn't?

    I agree with your experimentation argument. However we have had private schools for 2 decades in a bunch of countries. And while they are certainly not performing any worse than public schools, they are not "solving" the social problems that public school systems struggle with either.

    I would say a slight improvement is the upper bound on the empirical estimate on school choice.

    ReplyDelete
  47. Rodney,

    "I think you are sorely mistaken about race and culture being equivalent."

    That is not what I wrote. I wrote that one is empirically an sub-category of the other.

    If you are incapable of or unwilling to read what I write and understand it, you will have to forgive me for not taking your objections more seriously or being nicer to you.

    "Regarding statistics, of course you're going to get "significant" differences at very high sample sizes."

    As a good scientist, do you know what "economically significant" means?

    "And presenting the distribution of these scores (beyond the standard error that you reference) is rather meaningful."

    I didn't reference the standard error, I referenced the standard deviation. In the original text.

    Honestly, I am not a good statistician, by any means. However what I am doing is so simple that you don't need advanced statistics.

    However, having someone who doesn't understand the minimum of what I have written or the difference between standard error and standard deviation criticize my "science" and asking me to "hit the statistics books."

    When you write that, readers who don't understand assume that you know what you are talking about, which you clearly don't.

    As for the standard deviation, since you are incapable of doing it yourself, try this.

    Click here:

    http://nces.ed.gov/pubs2008/2008016.pdf

    Scroll down to page 45.

    Look under Heading: "Standard deviations of the average scores of 15-year-old students on combined science literacy scale"

    France 102
    Germany 100
    Sweden 94
    US 106

    Notice that the difference in variance across countries are small in magnitude compared to the differences in means. Notice also that they are unsurprising in their tendencies. Notice that the figures support my argument (being low in small demographically and culturally homogeneous countries).

    Notice last that information could for our purposes perhaps be summarized as writing

    "A crude rule of thumb here is that 50 points is 0.5 standard deviations."

    ReplyDelete
  48. Regarding African-Americans:

    1. I never wrote that they were "immigrants". Simply that they were not from Europe.

    I wrote clearly my method (I remove "all other groups than non-Hispanic whites), and my motivation (they are exclusively from Europe, everyone else is from somewhere else).

    2. Lots of empirical research has been done about African-American students scoring less than non-Hispanic whites.

    My reading of the current consensus is that:

    A. We have no evidence that it's due in any way to any genetic differences (the gap keeps declining following black social progress, for one)

    B. We have no evidence that it's due to the schools itself (the gap remains once school quality is controlled for).

    Here is a fairly typical paper in a good journal, by Fryer and Levitt:

    http://ideas.repec.org/a/oup/amlawe/v8y2006i2p249-281.html

    From the abstract:

    " Black children enter school substantially behind their White counterparts in reading and math, but including a small number of covariates erases the gap."

    "None of the explanations we examine, including systematic differences in school quality across races, convincingly explain the divergent academic trajectory of Black students."

    The University of Chicago view is that the gap is due to human capital investments. This includes parents and peer groups, and there is overwhelming evidence that black children have worse home environments than white children.

    Once you accept this evidence, it is a simple matter of understanding causality that you can't automatically conclude that the school system in America is worse.

    You can still conclude that American society or social safety net is worse if you want. But not the school system.

    More...

    ReplyDelete
  49. Let me explain in painstaking detail why it makes no sense to conclude European schools are better based on comparing averages, so there is no ambiguity about my methods and motivations:

    In the United States, socially privileged groups such as non-Hispanic whites score about 70 points more than disadvantaged minorities, by which I mean the average of blacks and Hispanics. The disadvantaged groups constitutes more than 35% of kids.

    In Europe, socially privileged groups such as native born Europeans score about 50 points more than disadvantaged minorities, by which I mean immigrants (mostly, but far from entirely, from the Balkans, The Middle East and Africa). The disadvantaged group constitutes less than 10% of kids.

    If Western Europe had solved it's problem of providing minorities with the same education outcomes as native white Europeans, what I would be doing would be highly questionable.

    This especially if we were evaluating the economic model of Welfare State Europe and Free Market America as a whole, and not just narrowly comparing the public school systems.

    However, Western Europe is getting a higher mean not primarily because they are better at helping their minorities, but simply because they still have very fewer minorities than the United States.

    It would be pure insanity to allow Europeans to continue attribute the differences in average outcomes to Europeans them a superior education system or being smarter than Americans, when we can plainly see that their education system is just as incapable of solving home problems as the American system.

    This is plain in Malmö in Sweden, and the suburbs of Paris.

    As long as France and Scandinavia only had a homogeneous population, they destroyed the United States and other countries in social outcomes. Witness the superiority of the European "Social Model"!

    As soon as their population mix changed, the social problems that according to liberal theory were "solved" by the welfare state, emerged, often worse than the U.S.

    Someone mentioned Caribbean's in the United Kingdom.

    The majority-minority test-score gaps in the U.K students is only a little smaller than the majority-minority gap in the United States (and larger than the gap between white Americans and black immigrants).

    http://www2.warwick.ac.uk/fac/soc/cedar/staff/stevestrand/strand_inpress_tiering_prepub.pdf

    It's just that the minority in question is 1% in the U.K, so the average result of the school system isn't influenced too much.

    But are we really supposed to accept that Europe would do better if (or when) they are 35% minority? They can't even deal with the difficulties a minority of 1% better than the United States.

    The liberal utopia of Sweden, the pride of Social Democratic Europe, couldn't even handle 50.000 Middle Easterners and Bosnians in Rosengård without collapse in all social statistics, including achievement scores. This is the exact same school system that the rest of Sweden has, and that liberals in the U.S admire so much.

    If I observe that the American school system does better with Europeans as the European School System, and no worse with low income minorities, why would I conclude that the school system is worse just because the composition of the two groups differs?

    ReplyDelete
  50. Race and culture are not empirically sub-categories of one another. That is absolutely ludicrous -- your race doesn't determine your culture, nor vice-versa. They correlate based on global distributions of races within cultures, but they're not the same thing at all.

    I'll assume "economically significant" means "meaningful effect size" in your field? Regardless, you're simplifying something so grossly as to make the stats that come out in the end meaningless.

    Since you reported SD globally rather than by group, and didn't really go into what you did, you'll have to forgive my confusing it for you reporting SE.

    If you think what you're approaching here is easy enough to be solved by slamming some means onto a chart, you're in the wrong. You've basically boiled an entire field down to, as you put it, undergraduate-level statistics. Using undergraduate-level presentation.

    Defend your science with science, not sarcasm. I don't care how rude you are; I do care to see whether what you're doing is meaningful or not. And right now it looks like you've worked with enough to be dangerous but not enough to approach the complexity of the topic at hand.

    ReplyDelete
  51. I must stress this again: Using a demographic variable like race rather than true cultural variables (RWA, fatalism, collectivism, etc.) or socioeconomic variables makes this whole exercise a bizarre approximation of comparing based on something meaningful.

    Your logic is that, for example, an African-American is best compared to someone from their "home country." What is their home country to you? Because in America, their cultural background is likely to be rather American.

    In short: Busted categorization variable, insufficient explanation of the logic behind what you're doing.

    ReplyDelete
  52. This comment has been removed by the author.

    ReplyDelete
  53. "in America, their cultural background is likely to be rather American."

    It is certainly American.

    Is it the same as non-Hispanic whites in America?

    And is it possible for America to have more than one cultural group?

    ReplyDelete
  54. 1. "Since you reported SD globally rather than by group, and didn't really go into what you did, you'll have to forgive my confusing it for you reporting SE."

    Why? Did that suddenly make you illiterate?

    Were you confused by the phrase "standard deviation"?

    If you are so unfamiliar with education statistics that you don't know that the variance is similar across developed countries, why not take a few minutes to actually learn the details of the subject?

    Next time, you might want to do that *before* you embarrass yourself.

    2. "race and culture are not empirically sub-categories of one another. That is absolutely ludicrous -- your race doesn't determine your culture, nor vice-versa."

    You are making two completely unrelated statements, which you seem to think are causally linked.

    Sure, race does not "determine" your culture, and culture doesn't "determine" your race.

    However, due to SPECIFIC historical reasons, the culture of the SPECIFIC demographic group African-American is different from the SPECIFIC group of non-Hispanic whites.

    A fact that matters in the SPECIFIC comparison of Europeans groups that I am doing.

    There are sometimes large overlaps in culture. But few non-Hispanic whites would best be characterized as belonging to the African-American cultural group. That is just an empirical regularity, having to do with a history of racial segregation in the United States.

    You have offered no real counter-argument, expect 2-3 incorrect claims that my statistics were wrong, some sort of emotional aversion to comparing groups, and an obtuse refusal to accept that race and culture are empirically linked in the United States.

    3."I'll assume "economically significant" means "meaningful effect size" in your field?"

    So you didn't know? That's fine.

    When I don't know what someone who is debating with me means, I look it up.

    I took the time to respond to you specific critique of the differences in averages not being large. I explained why you were wrong, in detail.

    Not only were you too ignorant to even understand what I was writing, you are too lazy and inconsiderate to even look it up.

    This didn't stop you from repeating a second time (and now a third time) that the differences in mean between Europe and the United States are somehow not large enough to be relevant. You are wrong.

    You can repeat it ten times more, each time revealing another part of the analysis you don't understand.
    Nothing will change.

    Either put the effort to first understand what I have done, and formulate why it's wrong. Or just be honest and admit that your problem is you just don't LIKE my conclusions and comparison, not that you have substantive criticism to offer.

    This way I can spend my time dealing with the issues raised by the people who actually read and understood what I did.

    ReplyDelete
  55. First off, you are putting words into my mouth. I never once said that the differences were not large enough to be relevant. What I continue to say is that your understanding of cultural variables is way off. You can't simultaneously impugn my reading comprehension and then make things up.

    Because of that ambiguity of variables, your results are quite misleading. What constitutes an "African-American" cultural group compared to a "non-white Hispanic" cultural group? In America, the main difference you will find is likely to be socioeconomic rather than based on some sort of substantive cultural variable.

    You're not truly correcting for culture but rather simply comparing racially. Which is not particularly sophisticated and lumps in a lot of variance that you're now mis-attributing.

    You can continue to insult me, but you do yourself no favors in making your arguments any more cogent. You've successfully compared European-background whites to European-background whites in different Western cultures. How does this get the field anywhere? You address the problem by removing it from the examination.

    ReplyDelete
  56. 1. "I never once said that the differences were not large enough to be relevant."

    Yes you did, even if you were striking out wildly at all directions, and perhaps not really clear yourself about what your objection was.

    Since you also were gracious enough to do this while calling me an idiot and creating the false impression that what I did was somehow statistically incorrect, I will not let you off the hook.

    I wrote:

    "the large differences in means we observe are apparently statistically and economically significant."


    You replied: "Regarding statistics, of course you're going to get "significant" differences at very high sample sizes."

    Even though I provided you with data showing that it is both statistically and economically significant. And even though I in addition took the time to explicitly spell this out for your (as everyone other commentator seems to have understood) specific benefit.


    2. "What constitutes an "African-American" cultural group compared to a "non-white Hispanic" cultural group?"

    Self-identification.

    3. "In America, the main difference you will find is likely to be socioeconomic rather than based on some sort of substantive cultural variable."

    What's your evidence for this confident claim? That may be conventional wisdom among sociology-undergraduates and among daily Kos readers.

    But social-science research established decades ago that factors such as income, while not irrelevant, are weaker explanatory variables than race.

    Just go down to page 13 of this report:

    http://www.mckinsey.com/app_media/images/page_images/offices/socialsector/pdf/achievement_gap_report.pdf

    The poorest whites have about the same average as the richest blacks.

    If variation in socio-economic factors was the main driver, non-Hispanic white Americans would not score higher than Scandinavia. There are plenty of poor whites. The poverty rate of non-Hispanic white Americans, both relative and absolute, is higher than Scandinavia.

    http://nces.ed.gov/surveys/pisa/idepisa/


    4. "Which is not particularly sophisticated"

    Simplicity in methods is an advantage

    5. "You've successfully compared European-background whites to European-background whites in different Western cultures. How does this get the field anywhere?"

    It suggest that the American education system performs better in the task of training Europeans, contrary to conventional wisdom.

    ReplyDelete
  57. 6. "You address the problem by removing it from the examination."

    My comparison generated strong and direct evidence when comparing European-Americans with Europeans, and Asian-Americans with Asians, but no direct evidence about comparing African-Americans with American-Whites.

    So naturally I wrote a lot about what the analysis generated evidence on, and not a word about what I didn't generate evidence about and was not trying to measure.

    You are the one obsessed with the implications of my comparison on racial differences within the U.S, to such an extent that you appear incapable of absorbing anything else.

    Since race is obviously all you care about, let me tell you about race:

    As Europe is becoming less homogeneous, it is becoming apparent that their social system is no better than the United States in terms of achieving majority-minority equality in school outcomes.

    This is apparent if we look at all immigrants, or if we look at particular groups such as Middle Easterners in Sweden or Caribbean Blacks in U.K. The majority-minority gap is similar to the one in America, it's just that they have too few minorities to alter the averages that newspapers report.

    The large number of liberal Americans who assume, based on Utopian notions of what things look like in Europe, that Europeanizing the United States will solve racial problems are thus using flawed inference.

    ReplyDelete
  58. Simplicity in methods paired with an overly reductionist approach is not a virtue.

    You're arguing against things I am not saying and ignoring swaths of variables that could mediate or moderate what's happening. Using self-identified race as an indicator of culture is simply bizarre. Don't call it what it isn't -- you're just comparing based on racial demographics.

    You'll also note that I never called you an idiot, only your logic and methodology poor and your statistical presentation suspect.

    Continue tilting at windmills, I don't think I can get through. You're glibly assuming away problems.

    ReplyDelete
  59. "ignoring swaths of variables that could mediate or moderate what's happening."

    So you are no longer defending any of your original points of criticism. You are now instead referring to unspecified "swaths of variables", to prove me wrong where you failed.

    I guess after many posts we have established that you don't need actual arguments to defend your beliefs, just the hope of the existence of potential arguments out there somewhere.

    So lets agree. It is certainly possible that some potential argument you have failed to think of proves me wrong. I hope you feel better knowing that.

    ReplyDelete
  60. As for Asians:

    The original comparison is between North-East Asians (in Japan-Korea-Hong-Kong and Singapore) with all Asian Americans. But only a little less than half of Asian Americans are Japanese-Korean-Chinese, and PISA doesn't break out these from all Asians.

    So let's look at another dataset just to compare Asian Americans from North-East Asia to the average for all Asian-Americans.

    Look at the NLYS97, a large representative dataset managed by the Department of Labor. The NLSY reports PIAT math scores and also report the country of origin.

    What I find is that Americans from North-East Asia (Japanese-Korean-Chinese) score about 0.2 standard deviations higher than the mean for all Asians.

    This is a fair deal, around 20 PISA points. The result is the same if other achievement tests included in the NLSY97 are used.

    This suggests that the advantage of North-East-Asian-Americans in PISA compared to actual Asia is almost certainly significantly greater than the original comparison suggests.

    On the other hand, as I wrote and others pointed out, Asians are positively selected. We are comparing a group of Asian-Americans selected partly based on their parents education performance with all kids in the 4 NE-Asian countries.

    These two biases go opposite directions, and it is hard to know which one is more important. I think they are approximately of similar size, based on estimates I have seen, with the differences between the 2 Asian groups perhaps a little more important than selection.

    But at least we now know approximately the size of one of the biases, and can do more than just speculating.

    ReplyDelete
  61. As Europe is becoming less homogeneous, it is becoming apparent that their social system is no better than the United States in terms of achieving majority-minority equality in school outcomes.

    This is apparent if we look at all immigrants, or if we look at particular groups such as Middle Easterners in Sweden or Caribbean Blacks in U.K. The majority-minority gap is similar to the one in America, it's just that they have too few minorities to alter the averages that newspapers report.


    Most of the Caribbean blacks in the U.K. are not recent immigrants. As I understand it, the population is well into the second and third generations.

    The recent-immigrant blacks in Britain are mainly from Africa, and with the notable exception of the Somalis they do quite well academically. One recent educational report showed African immigrant girls outscoring native British boys.

    Peter

    ReplyDelete
  62. "One recent educational report showed African immigrant girls outscoring native British boys."

    And you didn't see any problem with comparing boys from one group to girls from another?

    Here are I believe those statistics:

    http://www.statistics.gov.uk/cci/nugget.asp?id=461

    http://www.education.gov.uk/rsgateway/DB/SFR/s000564/

    The share of students who get good grades:

    White British (52%)
    Black (39%)
    -Black African (43%)
    -Black Caribbean (36%)

    Notice: A large racial achievement gap.

    Or read this article about the U.K:

    http://www.guardian.co.uk/education/2010/jun/15/black-students-less-likely-first-class-degree

    "Black students are more than three times less likely to be awarded a first-class university degree than their white classmates, a major study has revealed....

    Only 37% of the black students achieved a first or a 2:1, compared with 62% of the white students"

    I don't know which of U.S or U.K has the bigger education gap exactly, but it seems safe to say the problem is large under both systems. It's just that you are less likely to hear about it when it happens in Europe.

    In fairness, the U.K is not exactly a model of welfare state Social Democracy, but something in between.

    ReplyDelete
  63. ""Sweden had a lot of immigration in 50s and 60s from poorer countries."

    Nope. This is a myth. Sweden had no quantitatively relevant immigration from outside of Europe until the 1980s."

    I say "immigration", you say "immigration from outside of Europe".
    I don't really see why turks born on the "right" side of Bosphorus should outperform turks born on the "wrong" side and neither do I see why grandchildren of Yugoslavian immigrants should outperform grandchildren of Turkish immigrants(given your assumptions about culture).

    "I don't need data on all American groups, if I have data on all European countries of origin and weight accordingly. That takes care of your "smart German" problem."
    That's true only if you assume that taking Finnish schools and moving them to Greece would automagically make Greek scores jump 70 points. If you on the other hand believe that there are cultural differences that affect performance and those cultures change over time, then changes in, for example, Italian American and Irish American culture over time could very well be a plausible explanation.

    ReplyDelete
  64. ironrailsironweights--

    The recent-immigrant blacks in Britain are mainly from Africa, and with the notable exception of the Somalis they do quite well academically. One recent educational report showed African immigrant girls outscoring native British boys.

    This is garbage (Guardian pundit type) thinking and arguing, unsupported by links to studies or even articles discussing them. As Tino's detailed reply with links demonstrates very well.

    ReplyDelete
  65. An obvious short term policy implication for America is to stop further depressing average level PISA school attainment scores, by letting in or winking at the illegal immigration of groups that score way low on average on PISA and IQ tests, and whom we haven't screened for only the high smart/high IQ ones (as Canada and Australia mostly do, outside of asylum/refugee immigration). I.e., stop such mass immigration from Haiti, Mexico and Central America (the ones who come here especially illegally tend to be from the heavily indio, lowest levels of those societies), and no the third and 4th generations here DON'T do much better in e.g high school graduations rates.

    This is looking at the facts dispassionately, rather than from a politically correct, ideological filter.

    ReplyDelete
    Replies
    1. First of all, I appreciate the perspective in this article. I think any statistician will tell you that sometimes statistical means or "averages" can be misleading, and that oftentimes it is important to look at the range and standard deviation of a data set for a better picture. Still, athough 65% of students are white, non-immigrants, this still leaves a sizable portion of the population that may or may not be recent immigrants (as in the case of Blacks and many hispanics, and even Asians, at this point) who are graduating from high school and entering the work force with sub-par skills. Although your unique perspective is very compelling, it is definitely a skewed (to borrow a statistician's term and put my own spin on it) picture. Out of curiosity, I googled the nationalities of Nobel prizes, and found that the USA has more than any other country, including the UK, France, and Germany. A variety of factors are most definitely influences here.

      Delete
  66. OneEyedMan--

    OneEyedMan

    Without denying that that parental culture and educational attainment have some effect on educational outcomes, we can't let badly run urban schools off the hook that easily.

    A lot of why they're often badly run is the combination of:

    1) extensive class disruption by black and Hispanic/Amerindian students, mostly males; and

    2) the difficulty in controlling and ending this through various sorts of disciplinary measures, because of heavy and ever increasing left / PC restrictions on same, and also because of "disparate impact" lawsuit fears and actuality.

    See:

    http://isteve.blogspot.com/2010/12/pisa-forever.html

    ReplyDelete
  67. As I wrote, many Asians are children whose parents were selected based on ability. On the other hand, and as I again wrote; only half of Asians-American are from North-East Asia (Japan-Korea and various Chinese).

    Actually I don't think it's true of many NE Asians. It is true of most Indians. Most Japanese came here as agricultural laborers and most Chinese as laborers and often illegally. Not sure about Koreans.

    ReplyDelete
  68. Marco Brown-

    (Aside: data in America demonstrate black-African immigrants score better on educational tests and have better educational and vocational outcomes than black-Americans. This might be a result of selection bias, but it is what it is.)

    That's definitely the result of cherry picking/selection bias on average. One major route for African immigrants into America is a university student visa, and then either getting employment which makes them eligible for a green card, or visa jumping. Weighing on the other side of this generally are the African refugees we take, but the overall cherry picking effect is there. It operates even more strongly with Indian immigrants.

    ReplyDelete
  69. Marco Brown--

    One difficulty I have with this analysis is why certain groups (specifically black-Americans) are excluded from the European-origin group. The answer may appear obvious (they're of African descent), but that shouldn't necessarily exclude them from the European group. The majority of modern Black-Americans came here centuries ago as slaves. Since that time, they have become less and less African and more an more European. Much of predominant black-American culture still has trappings of Africanism, but they are much more like white Americans than black Africans. Hence, excluding them from the European-origin group and placing them in the African-origin group seems incorrect as a matter of methodology. I may be completely off base here. Please tell me if I am.

    This so massively fails any kind of common sense test that it's difficult to know where to begin.

    In fact I don't believe I've ever seen anyone try to argue before, when separating Americans out into broad groups or ethnicities, that African Americans belong in the same group as European origin Americans. Certainly those who support affirmative action, "minority" set asides (but not for the more successful minorities), and "disparate impact" tests for supposed discrimination don't do that.

    ReplyDelete
  70. Tino--

    Steven Sailer has done your same analysis not just comparing Euro origin American with Euro countries scores on PISA, but also comparing American NE Asians, Hispanics and Africans, with the PISA scores in their homelands (to the extent they participated in the PISA), and found that American schools did well for these Americans as well, compared to the schools in their ancestral homelands.

    Net, net, the reason that American schools look kinda bad on PISA is that about 35% of American school kids are NAM minorities (blacks and heavily indio Hispanics), which is only partly offset by the 5% or so who are generally outperforming Asian minorities.

    We aren't letting in tons more blacks, but we are letting in legally and illegally tons more heavily indio Mexicans and Central Americans, to the point where they are projected by the census bureau on present trends with no changes in immigration policy or enforcement, more numerous than Euro origin Americans by 2030, and earlier than that in our primary schools.

    ReplyDelete
  71. In America, the main difference you will find is likely to be socioeconomic rather than based on some sort of substantive cultural variable.

    Wrong. Endlessly repeated leftist horse twaddle.

    The SAT scores of high school students whose (self identifying) African American parents are in the top quartile income group are LOWER that the SATs of white students whose parents are in the lowest income quartile.

    Whites from households in the lowest income bracket have higher IQ scores than blacks from households in the highest income bracket:

    One of the most disturbing, I think perhaps the most disturbing fact in our whole book is that black students coming from families earning over 70,000 are doing worse on their SATS, on average--it's always on average--than white students from families in the lowest income group. You want to cry hearing that figure. I mean, it's so terrible.


    http://www.gnxp.com/blog/2007/10/james-watson-tells-inconvenient-truth_296.php

    It's obvious that these findings don't come from social scientists than wanted to come up with these results.

    ReplyDelete
  72. As Pat Buchanan recently said on this PISA scores by race across nationalities topic:

    Among the OECD members, the most developed 34 nations on earth, Mexico, principal feeder nation for U.S. schools, came in dead last in reading.

    Steve Sailer of VDARE.com got the full list of 65 nations, broke down U.S. reading scores by race, then measured Americans with the countries and continents whence their families originated. What he found was surprising. [PISA Scores Show Demography Is Destiny In Education Too—But Washington Doesn’t Want You To Know, December 19, 2010]

    Asian-Americans outperform all Asian students except for Shanghai-Chinese.

    White Americans outperform students from all 37 predominantly white nations except Finns, and U.S. Hispanics outperformed the students of all eight Latin American countries that participated in the tests.

    African-American kids would have outscored the students of any sub-Saharan African country that took the test (none did) and did outperform the only black country to participate, Trinidad and Tobago, by 25 points.

    America's public schools, then, are not abject failures.

    They are educating immigrants and their descendants to outperform the kinfolk their parents or ancestors left behind when they came to America. America's schools are improving the academic performance of all Americans above what it would have been had they not come to America.

    What American schools are failing at, despite the trillions poured into schools since the 1965 Elementary and Secondary Education Act, is closing the racial divide.

    We do not know how to close the gap in reading, science and math between Anglo and Asian students and black and Hispanic students.

    And from the PISA tests, neither does any other country on earth.


    http://www.vdare.com/buchanan/101227_pisa_and_bad_students.htm

    ReplyDelete
  73. You’re methodology is just as biased as the raw unadjusted PISA score.

    You basically took the EU and removed only the recent immigrants and their children. Then you took the US and removed much more than the recent immigrants and their children. You removed all the ethnic non-white minorities, many of whom with disadvantaged social status, including the ones who’ve been in the US for many generations. Hardly a comparison of apples to apples.

    Even more outrageously Eurocentric the comments by Doug, who hates the presence of so many low scoring individuals of Amerindian origin in America, maybe forgetting they were in the Americas much longer than the Europeans.

    Well, like it or not there is a reality you have to face: the US has to compete with the rest of the world with the population it has, not with the population some of you wished it had.

    ReplyDelete
  74. robertogoretti--

    who hates the presence of so many low scoring individuals of Amerindian origin in America, maybe forgetting they were in the Americas much longer than the Europeans.

    So what? They hardly made much of the place.

    But the one's who been in the US for ages are one thing; I'm against letting so many low scoring ones you came here illegally from other countries like Mexico and those in Central America remain, and tons more follow on illegally.

    To think that's good for the country's future is idiotic and ignores past experience with e.g. Mexican American high school graduation rates unto the fourth generation here.

    ReplyDelete
  75. You removed all the ethnic non-white minorities, many of whom with disadvantaged social status, including the ones who’ve been in the US for many generations. Hardly a comparison of apples to apples.

    You want an apples to apples comparison. Here's one:
    The analyses of the General Social Survey data from 1974 to 2000 replicate earlier findings from the National Longitudinal Survey of Youth that racial disparity in earnings disappears once cognitive ability is controlled for. The results are robust across many alternative specifications, and further show that blacks receive significantly greater returns to their cognitive ability than nonblacks. The trend data show that there was no sign of racial discrimination in the United States as early as 1970s. The analyses call into question the necessity of and justification for preferential treatment of ethnic minorities.

    So we see that many who are disadvantaged are not disadvantaged because society is discriminating against them. Most of the systems that Tino compared the US to don't have large segments of their population who are academically underperforming across multiple generations despite those generations being native born. That's a pretty big elephant in the room and any study that wants to look at the efficacy of the educational process in the international realm needs to address this issue in the study design.

    ReplyDelete
  76. the US has to compete with the rest of the world with the population it has, not with the population some of you wished it had.

    That's immaterial to Tino's research. He set out to compare educational process with student demographics being controlled, so that he could take a look under the hood and see how well US educational practices compared to those of other societies. It's the effectiveness of the process, rather than the effectiveness of the overall outcome, that is of interest here.

    Your observation that we have to compete with the population we have is true, but it's completely irrelevant to this focus of this study.

    What Tino has shown is that, when compared to other OECD nations, our educational process is producing good results with standardized inputs. The critics of US educational process want us to produce better results with diverse inputs (students) and for us to achieve that outcome we can't look for innovative techniques elsewhere in the world because no one knows how to achieve that desired goal.

    ReplyDelete
  77. Tangoman--

    The analyses of the General Social Survey data from 1974 to 2000 replicate earlier findings from the National Longitudinal Survey of Youth that racial disparity in earnings disappears once cognitive ability is controlled for. The results are robust across many alternative specifications, and further show that blacks receive significantly greater returns to their cognitive ability than nonblacks. The trend data show that there was no sign of racial discrimination in the United States as early as 1970s. The analyses call into question the necessity of and justification for preferential treatment of ethnic minorities.

    Yes, I've seen this too.

    Yet this information is hardly allowed by our mainstream media to get much play.

    ReplyDelete
  78. Tino, excellent post.

    I do, however, have to respond to this:

    "Similarly, the left claims that the American education system is horrible, because Americans don’t invest enough in education. The left has no answer when you point out that the United States spends insanely more than Europe and East Asia on education. According to the OECD, the United States spends about 50% more per pupil than the average for Western Europe, and 40% more than Japan."

    I have an answer: a large part of that expenditure in the US includes providing health care and retirement benefits to teachers and other public school workers. But in every other OECD country, those expenses are provided to all citizens nationally. So the US has to report those expenses on human welfare as part of its education spending, while other countries do not. That makes a huge difference.

    I am still waiting for some academic to do a comparison on spending that takes this into account. Tino, have you started on you dissertation yet?

    ReplyDelete
  79. Sorry to burst the bubble of all the praise here, but this analysis is generally meaningless.

    Academics can always hold one variable constant, or "correct for demography" to obtain a result or illustrate a picayune point.

    This is fine as far as it goes. No one reading this post should be left with the illusion that Shanghai is representative of all China.

    All of that said, nothing in this post should allow us pretend that the American education system is badly failing the nation.

    Where it is "succeeding," it does so at an extremely high cost, and for the urban and rural poor, it is nothing short of a moral outrage.

    At some point, we have to look beyond futile exercises in "correcting for X," and calculate the true cost of leaving the vast majority of Americans either a) less educated than they should be, or b) unable to understand the operation of the world around them (from science to simple economics).

    While we are calculating that cost, we need to compare our schools with the education outcomes that we SHOULD or COULD have, and stop measuring for the number of "corrected factors" that can fit on the head of a pin (or in a white paper).

    America spends more on education (save Switzerland, perhaps), and gets less, than any other nation. Also, it isn't unnoticed that many states can no longer afford to shower a generally mediocre class of bureaucrats and teachers with unwarranted pay and benefits.

    Where should we look for improvement? Specifically, we need to enact a basket of policies that transform (not reform) education.

    These include, but are not limited to;

    1.Empowering citizens who are unhappy with their schools to close and/or convert them to better schools.

    2. Start the process of having the money follow the child to a much more dynamic basket of schools and/or content providers.

    3. Create a high set of content standards, and mandate that those standards are measured OUTSIDE the control of the providers.

    4. Abolish/Dismantle the expensive, and completely worthless entity called a "district" and empower principals and teachers to meet higher standards in various ways.

    Lastly, this post, like almost all education commentary, focuses on comparing systems, when it is the concept of systems (for America, anyway) that need to be questioned.

    We need to stop caring about the pay, benefits, and contracts of the closed class of protected education jobs, and start caring about making sure that every child has a shot at a good education.

    That is why, despite it's well-thought out number crunching, this post misses the mark. Instead of "correcting for X" to make things look better, we need to get X into the kind of classroom that works for them. This can NEVER be accomplished in the context of the existing education system. This is why it needs to be dismantled.

    ReplyDelete
  80. Duke--

    I have an answer: a large part of that expenditure in the US includes providing health care and retirement benefits to teachers and other public school workers. But in every other OECD country, those expenses are provided to all citizens nationally. So the US has to report those expenses on human welfare as part of its education spending, while other countries do not. That makes a huge difference.

    If an academic were to correct for those things he should also correct down to after tax salaries, since Euro taxes are generally higher. VAT (a kind of sales tax not broken out from the price of an item), which is huge at 18% and more in many Euro countries, would also need to be considered.

    Such higher taxes are after all how the Euros pretend to pay for their universal publicly funded health care and pension systems.

    ReplyDelete
  81. Is PISA "a Sputnik wake-up" or are international comparisons invalid. Rather than wade into that debate, I'd rather look more closely at the questions in the PISA test and what student responses tell us about American education. You can put international comparisons aside for that analysis.

    Are American students able to analyze, reason and communicate their ideas effectively? Do they have the capacity to continue learning throughout life? Have schools been forced to sacrifice creative problem solving for “adequate yearly progress” on state tests?

    I focus on a sample PISA question and an insight into what American students can (and cannot do) in my post "Stop Worrying About Shanghai, What PISA Test Really Tells Us About American Students" http://bit.ly/eChNoY

    ReplyDelete
  82. Doug, I disagree. The assertion is that the USA spends more than other countries to school their children. I contend that the USA reports costs as part of education spending that are not reported by other countries; that makes a comparison impossible.

    Whether school employees pay more in taxes, and whether they receive any benefit from those taxes - those are separate questions.

    Put another way: if you asked two pizzerias how much they spend to make a pizza, you wouldn't want one to include the price of cheese while another omitted it, would you?

    ReplyDelete
  83. Duke:

    This is an accounting exercise. Most benefits for teachers have to eventually be paid for in Europe too.

    So sure, you may want to add 25-50% to the cost of teacher salaries in the U.S.

    But typically in Europe, you have to add 25-50% to cover payroll taxes and the extra private benefits that teachers often have.

    Payroll taxes are than used to pay for pensions and other benefits. If you adjust for details in one country, you have to adjust for details in every country.

    In order to convince you, let's sidestep the accounting and look at how many teachers each country ultimately can buy:

    According to the OECD, 2008, Ratio of students to teaching staff in educational institutions, Primary education.

    The number is students per teacher (so the lower the better)

    24.1 Korea
    18.8 Japan
    16.2 EU-15
    15.4 Switzerland

    14.3 U.S
    10.8 Norway

    So yes, it appears that American does indeed buy more teachers for the money you spend.

    Let me also repeat:

    American teaches work more hours than Europe (cheap per hour).

    American teaches are on average better educated than Europe (cheap corrected for quality).


    My dissertation is on the blog, scroll down a little.

    ReplyDelete
  84. Tino, forgive me, but I'm not following your logic.

    I understand the benefits need to be paid for by European countries. The question is whether those costs are reported as education expenses in the OECD reports. My understanding is that health care and pension costs are reported as part of the USA's expenses, but not as part of the expenses of other countries.

    If that's the case, the the OECD comparison doesn't do us much good.

    I suppose the payroll taxes that employers pay is an issue: here in the states, the FICA tax is split between the employer and employee (6.2% each). Don't know how it works elsewhere, but that's the point: how are we to compare if the OECD data is self-reported and not adjusted to make apples-to-apples comparisons?

    To use you to make an analogy: if the PISA scores need to be adjusted to be useful and give us an accurate picture of relative student performance across the OECD, shouldn't spending figures be put into the same context?

    I hear there's some good analysis coming out this spring on just this topic - I'm sure to write about it on my blog:

    http://jerseyjazzman.blogspot.com/

    Until then: I agree your figures on student:teacher rations are compelling, but until we know the relative expense of a teacher in each country, I'm hesitant to say it proves the point that the US spends more. Further, it doesn't look like the ratios you give match up neatly with the OECD graph you use.

    Again, this is a great post, and I appreciate the rigor of your thinking; we need more people like you doing this kind of work.

    ReplyDelete
  85. Excellent post. I didn't read through every comment, so I apologize if someone already brought this up, but the real stunner for me was when the scores are adjusted based on economic figures.
    http://voices.washingtonpost.com/answer-sheet/research/how-poverty-affected-us-pisa-s.html

    The question Education Reformers ask is "why are our schools so bad?" when the real question should be "why do poor students do poorly in school?"

    The problem with American test scores exists solely in the Trailer parks and Inner-city Projects of America. We have a poverty issue and a single parent/poverty epidemic, but not an education one.

    ReplyDelete
  86. Your argument is not completely accurate. Test scores are terible in most inter city schools where all the students are english speaking. In Europe even the poor kids are getting a decent education. In america we know most affluent suburbs have good schools but its the job of teachers to educate all students. We pay TOP dollar in our country to educate our kids and I expect the schools to give every american kid a good education. No excures that the parents may not care as much in poorer neighborhoods. I have difficult times in my job and I am not allowed to make excuses. If you have to live in an expensive neighborhood to expect a quality education in this country then something is wrong. Watch the movie Waiting on Superman. This guy seemed to give poor kids better educations than most of the richer suburbs. Teachers unions are the problem because of tenure. How can you expect to movtivate a teacher when you cannot fire them and they all get paid the same no matter how they perfrom.

    ReplyDelete
  87. I personally don't like the American school system. I personally don't believe in a lot of those studies, because in my own experience I was miserable in the American school system. My brother, however, adapted to the American culture from day 1, and even though I always took more advanced classes, had a higher GPA, and scored better than him in tests, he ended up going to Cornell (pretty much all paid) and I went to a regular state college. School has always feel like a jail in this country to me, and even though I'm a very good, and dedicated student and worker, it feels miserable when you don't enjoy learning, it drains the motivation out of me.

    ReplyDelete
  88. GDP isn't the economy, it isn't the society, and it isn't the real world.

    Kirk Sanford Sightline Payments
    Irish lottery

    ReplyDelete
  89. Please do not comment on things you do not understand. Shanghai students are not the highest scoring students. This is evident from the yearly college entrance exam, where students from other provinces consistently outscore Shanghainese.

    Yes, Shanghainese are richer. But the fact that these non-Shanghai students are not as rich drive them to study harder in school.

    ReplyDelete
  90. "I'm a Singaporean of Chinese descent and most Singaporean immigrants in the US come from a much higher socio-economic stratum than the average Singaporeans. The same is probably true of Chinese (Taiwan, HK and PRC) immigrants to the US. They are in no way representative of the home populations in NE Asia. "

    Wrong. Singaporean immigrants to the US are a tiny insignificant portion of the East Asian American population. The vast majority of Chinese Americans (80%) are lower class economic migrants from wide range of eras. The Japanese are overwhelmingly plantation cheap labor. Koreans, at least 25% are related to lower class war brides. The lion's share of the rest are lower-middle class Koreans who cannot compete in Korea.

    ReplyDelete
  91. The majority of Chinese-American are first generation immigrants who came after 1965 when the quota for Chinese people was increased drastically. It has been found that post-65 Chinese immigrants from Taiwan/Hongkong/China had the highest educational level of all immigrant groups [1]. Migrating to the US takes considerable resources. You either had to have the money or the education. It would be misleading to say that Chinese immigrants belonged to the lower class economic migrant group.

    You need only look at 1st generation Nigerian-Americans as an example. There is an usually high concentration of college-educated people in this group. Intuitively, there is no way this group would reflect the SES profile of a developing country.

    [1] Portes, Alejandro and MacLeod, Dag(1999) 'Educating the second generation: Determinants of
    academic achievement among children of immigrants in the United States', Journal of Ethnic and Migration Studies, 25:
    3, 373 — 396

    ReplyDelete
  92. These "post-65ers" are a tiny minority of the Chinese American population- Taiwanese Americans as the largest group constitute less than 20% of Chinese Americans.

    ReplyDelete
  93. Please cite me a source that says that post-65ers are a tiny minority of the Chinese-American population.

    ReplyDelete
  94. http://en.wikipedia.org/wiki/Taiwanese_American

    Approximately 600,000
    About 0.2% of the US population

    And it's very bold, and arrogant, to assume that these Chinese Americans are "elite". Not a single one of Taiwan's big CEOs is in America.

    Recent arrivals from the mainland, aside from people smuggled in on shipping containers, don't even number 100,000- and they're mostly concentrated on universities.

    The majority of Chinese Americans are the dregs of China.

    ReplyDelete
  95. I was just going through the stats of Pisa and you really want to compare apples to apples, look at Canada. Similar culture and language but with a much higher proportion of new immigrants and refugees than the USA and many geographical disadvantages to boot. It still gets in the top 3 or 4 of the world. Imagine if you just cherry pick Caucasians only in Toronto, we would blow China out of the water. If they can pick their richest city that is almost 100% homogenous Chinese, why can't we.

    Its a bit pointless right?
    In my mind, the best education level is the one that lifts every student to excellence and 'Leaves no child behind' as in Canada.
    With the proper system, new immigrants beat the natives no problem.
    Asa new immigrant you should know this yourself.
    The best stat that blew me away was that Canada achieves these remarkable results despite being second to last (before Kazakhstan) in teacher salary after 15 years of experience vs GDP. Talk about getting bang for your buck. Again, socialism wins out.

    Cheers,

    Bob

    ReplyDelete
  96. "Take me as an example. The school me and my brother attended ..."

    How can someone who has not mastered basic English grammar comment on educational peformance?

    Hilarious: the writer of this article is a patsy and shill for the teachers unions who are defending their awful record. That's why Diane Ravich tweeted this nonsense.

    ReplyDelete
  97. It is worth noting that difference of learning a language is not usually looked at. Linguists know that some languages are harder to learn than others. All linguists know that English is a very hard language to learn (to read and to write) since it has a very irregular spelling system. Finnish is on the opposite end of that spectrum, being very predictable and regular. Even though Finnish students start school at age 7, they are the top students on PISA (international) tests. In Finland, Swedish-speaking students (who are from higher socio-economic background) are no match for the Finnish-speaking kids. For a full explanation, go to http://finnish-and-pisa.blogspot.com/

    ReplyDelete
    Replies
    1. Point taken (and this could be a huge factor in verbal PISA scores, especially in difficult langauges like French), but Finland also hold the number 1 place for math, and the number 2 place for science. A random association: Finnish people also have much higher than average digit ratios, something which was found to be related to math and analytical ability (see John T. Manning's book for more info). More random facts: Finland is also an ethnically homogeneous, socialist country where there is much less disparity in income, making poverty less of a factor influencing scores.

      Delete
  98. I have gained a lot of knowledge by reading your post. It is really a wonderful blog endow with a lot of knowledge for readers who wants to know about the subject matter. Thank you very much.
    Academic Writing

    ReplyDelete
  99. Best IT online training on all IT courses monstercourses.obiee,informatica,cognos,etc courses are providing and one free demo,placement assistance,materials,software installation,etc are providing by monstercourses.for more pl visit our website or contact:info@monstercourses.com

    ReplyDelete
  100. "For example, we don’t have any real equivalent to Advanced Placements classes. We have cheaper and worse textbooks. The teachers on average have far less education. I could go on."

    I'd put it like this instead; The US offers AP classes due to it's standard in Europe. That's why an American high school graduate needs to spend 4 years instead of 3 at college to get a bachelor degree. To go through an equivalent education to my Swedish technical gymnasium education in the US you'd have to set up an integrated program with majoring AP science classes all through.

    http://en.wikipedia.org/wiki/Talk%3AEducation_in_Sweden#American_comparison

    ReplyDelete
  101. You can argue as much as you want for which school system is the best but that won't raise any national PISA scores. Each country is unique and has to deal with their education system issues in relation to it's current conditions and social politics. Apparantly, the school system in the US isn't very well adapted to all it's people and needs an overhaul without any excuses. I think much could be done here, especially when it comes to the parent-student-teacher communication in poor areas with social problems. One size doesn't fit all you know. The most important thing is to make students believe in themselves by require them to constantly improve by having engaged teachers who really wants to teach.

    ReplyDelete
  102. "It's dumb to compare one of the most elite cities in a country with entire nations, and to draw policy-inference from such a comparison. Shanghai has 3 times the average income of China! It is also naive to trust the Chinese government when they tell us the data is representative of the entire nation. Either you compare Shanghai to New York City, or you compare the entire country of China, including the rural part, with other large nations. Most of the news and policy conclusions we read about PISA-scores in the New York Times is thus pure nonsense."

    Perhaps you should also make corrections in relation to the income levels among American with European ancestry in comparison to income levels among white Europeans in Europe? .. It's well known that wealthy families care to send their kids to either private schools or schools with very good reputations in much larger extent than other people. In the the US there are also alot more options if you have the money regarding this than in Europe. Also, I've read that many Asian-American families move to certain areas for the sole purpose to be able to send their kids to schools with great reputations. I think it would be more accurate and relevant to compare students with family backgrounds being roughly on the same corresponding wealth levels in each country attending corresponding school systems, be it either private or public or charter schools or whatever. You're comparing whole school systems after all.

    ReplyDelete
  103. Interesting piece of study, thank you!

    If you are keen and want to know more about the education of children of immigrants in Finland, there is a good University of Oxford Thesis available, made by Elina Kilpi.


    Elina Kilpi, (2010). The education of children of immigrants in Finland. DPhil. University of Oxford.

    ReplyDelete
  104. Iam really satisfy by your information.
    It's well-written, to the point, and relative to what I do.
    I like it very much for giving information on
    Excellent Informatica Online Training
    .

    ReplyDelete
  105. Tino. Wait a second here. Would you not agree that you also have to look which kind of immigrants you are comparing with natives? Maybe, if Canada recieves more highly educated immigrants (because of their brain drain system of picking the ''right'' immigrants), while Sweden receives more low skilled immigrants, this will affect your results? Or am I wrong on this?

    ReplyDelete
  106. nice post and I found something new and folks can get useful info about BEST ONLINE TRAINING

    ReplyDelete
  107. I want to appreciate your excellent work. Nodoubt your post is praiseworthy. This information is quite unique for me because I have very little information about this.
    Funny Jokes

    ReplyDelete
  108. In order for people to rely mainly on an income insurance in case of unemployment, their government should do drastic economic development measures.

    ReplyDelete
  109. Thanks for sharing as it is an excellent post would love to read your future post
    schools in vadodara

    ReplyDelete
  110. The grandfather by Don fabio
    The Grandfather, Don Pacio Luciano, a reasonable, honourable man of integrity had sought justice after the deadly attacks on his family by suspected members of the notorious Sicilian Cosa Nostra. This is an outstanding story of love, honour, vengeance and astonishing revelation of the secret operations......
    Buy from Here:>> http://thegrandfatherbydonfabio.blogspot.com/

    ReplyDelete
  111. This comment has been removed by the author.

    ReplyDelete
  112. Great Articles and Nice post! I have discussions with US conservatives and liberals often who want to excoriate the public school system and punish the 'bad' teachers through various means, as if the system is just awash with them.
    I am really satisfy by your information.

    Assignment help uk

    We are provide for homework help tutoring,
    essay writing help,
    Human Resource management help,
    Management help,
    students assignment queries,
    Students Assignment Help ,
    Help With Assignment to students.

    ReplyDelete
  113. One of the objectives of Santa Monica Canada Migration is to welcome skilled newcomers who will contribute to Canada's growing economy.  Skilled workers that settle in Canada on a permanent basis are especially valuable to the Canadian workforce. Over the years Santa Monica Canada Migration Program planning numbers have fluctuated according to the priorities and economic and political considerations of the government of the day.Santa Monica Canada Migration  serves as the foundation for continued economic growth and which brings people, customs and traditions, rituals and culture to the forefront of current Government policy.

    ReplyDelete
  114. Just another brainwashed "American" denying the facts. Move along folks, nothing factual to see here...

    ReplyDelete
  115. nice info
    tanks
    http://www.educationsworld.com/

    ReplyDelete
  116. Great post and tips! This is clearly valuable information.

    caravan memory foam mattress

    ReplyDelete
  117. A great post! More question like this needs to be asked. I can't agree with myself thought if you can really do a comparison in this way. You are excluding a much larger part of the American population than the European. So your premise is that decendants of Europeans in America would be equally well off if blacks, latinos etc were removed from the country. Perhaps, but I can see many problems with that hypothesis. If you conducted this study back in the time of slavery, it would have been obvious to anybody that the wealth and prosperity that whites in the South enjoyed could not be possible without black slaves. If you excluded those, the propserity of white southernerns would diminish.

    White American middle class has a lot of purchasing power gained compared to their European brothers due the very low price of common services due to low wages earned by blacks and latinos. This puts more money at their disposal for e.g. education.

    In comparison the poor immigrants in western europe are not as large benefit to the middle class there. Part of this is due to much higher minimum wages and more welfare goods for poor immigrants. Countries such as Norway and Sweden take in far more refugees than the US, which is a much larger drain on financials than people who come with skills to the US to work.

    Still I think what is really important about your work, is that it shows the importance of raising up those at the bottom of society. America can't beat asians countries by focusing exclusively on white middle class schools. America has to solve its poverty and inequality problem. As you say, it isn't primarily what is happing inside the classrooms which are keeping America behind.

    Btw I am a bit in doubt about your assertion that poor inner city schools get as much funding as that of middle class schools. Are you sure that includes all funds? In good neighbourhoods a lot of funds come from fund raising among the parents. Also as far as I understand schools are locally financed by taxes from the school district or similar area, so that it will be affected by the income of those living there.

    Second point is pre-school. Studies the last years show that what happens before school is very important. Poor neighbourhoods will not have access to the same quality pre-schools as rich. Parents would simply not afford it. It is not almost free child care in the US like in e.g. Nordics, France etc.

    ReplyDelete
  118. Great post, thanks for sharing such nice info that is really very beneficial, thanks again. essay writing service

    ReplyDelete
  119. YOURWAY Migrations was established in 2009 with the sole purpose of guiding individuals accomplishes their education and immigration goals with ease and without much ado. Initially we started with few main countries and coaching of foreign languages.

    ReplyDelete

Google Analytics Alternative